Summarise a bgam fit

Returns a summary.bgam S3 object containing the call, family, sample and predictor dimensions, boosting parameters (mstop, nu, df_target), CV-selected mstop_opt (when autotune = TRUE), bagging info (when n.boot > 0), training loss at mstop, residual sigma for gaussian fits, and a full predictor selection-frequency table sorted in descending order. The selection frequency is the fraction of boosting iterations in which each predictor was the active base-learner.

Usage

# S3 method for class 'bgam'
summary(object, ...)

# S3 method for class 'summary.bgam'
print(x, digits = max(3L, getOption("digits") - 3L), ...)

Arguments

object

An object of class "bgam", as returned by bgam().

...

Currently ignored.

x

A summary.bgam object returned by summary.bgam().

digits

Number of significant digits for numeric output. Default max(3L, getOption("digits") - 3L).

Value

An object of class "summary.bgam". Print it with print.summary.bgam().

Invisibly returns x.

See also

bgam(), print.bgam(), plot.bgam()

Examples

set.seed(1)
n <- 200
x <- matrix(rnorm(n * 4), n, 4)
y <- sin(x[, 1]) + 0.5 * x[, 2]^2 + rnorm(n, sd = 0.5)
fit <- bgam(x, y, mstop = 50, autotune = FALSE)
summary(fit)
#> Call:
#> bgam.default(x = x, y = y, mstop = 50, autotune = FALSE)
#> 
#> bgam summary:  family = gaussian 
#>   n = 200   p = 4 
#>   mstop = 50   nu = 0.1   df_target = 4 
#>   Training mse at mstop: 0.3297 
#>   Residual sigma: 0.58 
#> 
#> Predictor selection frequencies (sorted):
#>  predictor sel_frequency
#>         V2          0.64
#>         V1          0.36
#>         V3          0.00
#>         V4          0.00